Abstract
Purpose :
We previously presented the first method (using optoretinography) to objectively and non-invasively measure individual cones’ spectral sensitivities in human subjects [1]. Here, we compare our measurements from a cohort of six color-normal subjects to Baylor et al.’s (1987) electrophysiological data [2].
Methods :
We used the Indiana AO-OCT system to acquire 5-s videos at a 10 Hz volume rate of a 0.8°×1° patch of cones at 5° temporal retina (to avoid macular pigment) in six male color-normal subjects (21-25 y/o). AO-OCT cone optoretinograms (change in optical path length beweteen ISOS and COST) were measured after stimulation with bright flashes (105 – 520 µW at cornea, <2 nm bandwidth, 5 ms duration) of 434, 458, 473, 488, 514, 544, 568, 589, 607, 632, 646, and 661 nm light from a supercontinuum laser. Cones were classified into S, M or L types, and their peak responses to the stimulus were recorded for each combination of flash strength and wavelength. A power law was then fit (after compensating for lenticular absorption) to hundreds of cone responses to extract fitting parameters used to compute spectral sensitivity. Individual sensitivity curves were averaged and interpolated with cubic spline to find the peak sensitivity wavelength per cone type. These curves were then normalized and averaged across subjects to compare to electrophysiology.
Results :
Average AO-OCT spectral sensitivity values (N=6) were obtained for S, M, and L cones at 12 stimulus wavelengths totaling 36 measurements (not shown). Average peak sensitivities for S, M and L cones occurred at 435±2.6, 531±4.1, and 560±4.6 nm (mean±SD), respectively, and are consistent with Baylor et al.’s 430, 531, and 561 nm peaks [2]. Mean-squared errors between our curves and Baylor et al. for S, M, and L cones were 0.002, 0.006, and 0.007. While the dynamic range of our M and L cone measurements was < 2.5 log units, our S cone measurements exhibited a dynamic range of ~5 log units. If longer or shorter wavelengths are tested, we expect M and L cone measurements to have a similar noise floor.
Conclusions :
AO-OCT optoretinography permits the first non-invasive measurement of individual cone sensitivity in the living human eye that is consistent with ex vivo electrophysiology.
[1] Bernucci et al. (2022). IOVS 63(7), 397-F0435
[2] Baylor et al. (1987). Journal of Physiology 390(1), 145-160
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.